d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
37,368 Commits 4 Branches 49 Tags
Commit Graph

14813 Commits

Author SHA1 Message Date
peter
d6f4bd18f5 Use COMPAT_PCI_DRIVER() for registration if it exists. This shouldn't 
hurt the driver portability to 3.x too much for where drivers are shared.
 1999-04-24 20:17:05 +00:00
peter
e7462e4ae5 Replace the pcidevice_set linker set based configuration mechanism for old 
style pci drivers with a simple one-line change to use a module that
registers itself under new-bus and should in theory enable just about all
of the pci drivers to be loadable (kldload and loader(8)) but without
having the impact of converting the APIs yet.

This also fixes the problem of having undefined variables when only
new-style pci drivers are present.
 1999-04-24 19:59:20 +00:00
peter
484e1a3ce1 Zap some #if 0 code for wired PCI devices. This will look quite different 
under new-bus.
 1999-04-24 19:55:41 +00:00
dt
3843abdc03 Fixed printf format errors on alpha. 1999-04-24 18:50:48 +00:00
phk
5bbdd11011 typo in comment 
PR:		11285
Submitted by:	Chris Costello <chris@calldei.com>
Reviewed by:	phk
 1999-04-24 18:26:54 +00:00
ache
b9837c617f so_linger is in seconds, not in 1/HZ 
PR: 11252
Submitted by: Martin Kammerhofer <dada@sbox.tu-graz.ac.at>
 1999-04-24 18:25:35 +00:00
kato
4891334365 Changed the type of id_port from short into int to avoid wrong 
conversion from short to unsigned long which is an argument of
bus_alloc_resource.  Since the value -1 is used to indicate no port
reousece, id_port need to be signed (suggested by Doug Rabson and
Peter Wemm.)
 1999-04-24 18:24:43 +00:00
ache
cfd84c8eac Lite2 bugfixes merge: 
so_linger is in seconds, not in 1/HZ
range checking in SO_*TIMEO was wrong

PR: 11252
 1999-04-24 18:22:34 +00:00
dcs
e62810c9b1 Minor nit. It is probably still incorrect, but at least don't 
claim to use i386-specific files on alpha...
 1999-04-24 18:05:45 +00:00
phk
711d7d880d Fix a braino in the v_id wraparound code. Give more (current) details 
in comment.

PR:		11307
Spotted by:	Ville-Pertti Keinonen <will@iki.fi>
 1999-04-24 17:58:14 +00:00
dcs
073326ad33 Move loader.conf.5, while it is still in it's first revision. 1999-04-24 17:49:47 +00:00
dcs
aa57b67a29 Add loader.4th man page (worst man page I ever wrote -- reviews 
are welcomed). Correct a nasty bug in loader.4th before anyone
notices, add a couple of features.
 1999-04-24 17:25:35 +00:00
dt
e45052d148 Fix spelling of '#if 0'. 1999-04-24 16:39:52 +00:00
peter
ddc40aed53 Drop tty/net/bio/cam interrupt class labels, it's meaningless here now. 1999-04-24 16:07:51 +00:00
peter
b4268d2f61 Drop the tty/net/bio/cam interrupt class labels, they are meaninless here 
now.
 1999-04-24 16:04:16 +00:00
dt
ce1c742794 Use pointer arithmetic as appropriate. 1999-04-24 13:23:48 +00:00
luigi
069fcdc17d postpone the sending of IGMP LEAVE msg to after deleting the 
mc address from the address list. The latter operation on some
hardware resets the card, potentially canceling the pending LEAVE
pkt.
 1999-04-24 12:28:51 +00:00
dt
9b8660ce53 Fixed printf format errors on alpha. 1999-04-24 11:29:48 +00:00
dt
f84bfd9862 Use pointer arithmetic to do pointer arithmetic. 1999-04-24 11:25:01 +00:00
dt
98cf21bf1e Modify the non-i386 version of disable_intr() and enable_intr() so that they 
don't produce a warning on every use.
 1999-04-24 10:41:21 +00:00
peter
92e2d421da Back out parts of rev 1.4; making isa.c preserve the hints across a 
release made it unneeded.
 1999-04-24 07:04:51 +00:00
peter
30825c6cf7 Revert part of 1.9; we don't need to reset the port after release now that 
the isa bus doesn't clear the hints at that point.
 1999-04-24 06:48:27 +00:00
peter
8d543f8095 Don't clear the hints on release, just the resource containers. 1999-04-24 06:47:24 +00:00
peter
57d6459b70 Typo, use SYS_RES_IRQ when releasing the irq resource, not SYS_RES_IOPORT. 1999-04-24 06:46:10 +00:00
kato
f0dd940a69 Optimize better for space. 
Submitted by:	Takahashi Yoshihiro <nyan@wyvern.cc.kogakuin.ac.jp>
 1999-04-24 04:29:13 +00:00
kato
4be4197e7b Commented out adv_isa.c. 1999-04-24 04:22:57 +00:00
kato
b2970ba31d - PC98 doesn't support 16bits bus I/F of adv deriver. 
- PC98 has the bs driver.

Submitted by:	Takahashi Yoshihiro <nyan@dd.catv.ne.jp>
 1999-04-24 04:21:42 +00:00
kato
fbfb955028 1MB is not 1024 * 1024 * 1024 but 1024 * 1024. 1999-04-24 04:16:22 +00:00
gibbs
8da13829f4 ahc_pci.c: 
Convert to new bus and bus dma.
	Use latest PCI API.

bt_pci.c:
	Fix a few bugs in how resourses are released left over from
	when this driver was converted to new bus.
 1999-04-23 23:30:22 +00:00
gibbs
6093db57d5 staticize. 1999-04-23 23:29:23 +00:00
gibbs
2aed7d023f ahb.c: 
Synchronize malloc types for our softc.

ahc_eisa.c:
	Sync to aic7xxx driver bus dma and new bus changes.
 1999-04-23 23:29:00 +00:00
gibbs
bc3f9921db bt.c: 
Simplify some inline functions.
btreg.h:
	Correct a comment.
 1999-04-23 23:28:20 +00:00
gibbs
405e2ed769 Complete conversion to bus dma. This driver now works on the alpha. 
aicasm_symbol.c:
	Correct an unaligned access problem.  You can't rely on DB to
	store your data in an aligned fashion.
 1999-04-23 23:27:31 +00:00
gibbs
998c74d380 Kill a compiler warning. 1999-04-23 23:26:10 +00:00
gibbs
4bdf5acb03 Release bus entries in the EDT when a path is released. 
Null out path entries when the path is released to catch bugs.
Free our configuration hook when we're done with it.
 1999-04-23 23:25:48 +00:00
peter
7bc6edd255 Make the register_intr() glue actually have a chance of working... 1999-04-23 21:01:19 +00:00
peter
1d952b2530 s/lkm(4)/kld(4)/ 1999-04-23 20:53:34 +00:00
dt
9efe75f7a6 Make pmap_collect() an official pmap interface. 1999-04-23 20:29:58 +00:00
dt
c270d3aded Moved cpu_set_fork_handler's prototype from <machine/cpu.h> to <sys/proc.h>. 
Suggested by:	bde
 1999-04-23 20:22:44 +00:00
dt
a1d50daf71 Fixed several (not all) warnings. 1999-04-23 19:53:38 +00:00
ghelmer
aa25704686 Reference kld(4). 
Submitted by:	Nathan Ahlstrom <nrahlstr@winternet.com>
 1999-04-23 16:12:44 +00:00
peter
a46cf5f8fa Update VM86 comment - it's used for VESA too. 
PR:		7976
Submitted by:	Stefan Eggers <seggers@semyam.dinoco.de>
 1999-04-23 06:30:10 +00:00
jdp
43e60106b0 Update #include statements to reflect the new location of "sioreg.h". 
With these changes plus the egcs fix I committed a few minutes ago,
"make -DWANT_AOUT world" works again.  Most likely, "make upgrade"
is fixed too, though I haven't tested that.
 1999-04-22 21:02:44 +00:00
davidn
8946b02fed Fix so that this driver works again when compiled with EGCS (which 
apparently has a much better optimiser, requiring 'volatile' memory
to be declared as such).
 1999-04-22 15:19:37 +00:00
peter
31282b93d0 Return the port size from the probe. 
"ppc0 at port 0x378 irq 7 drq 3 on isa0" becomes
"ppc0 at port 0x378-0x37f irq 7 drq 3 on isa0"
 1999-04-22 13:10:43 +00:00
sos
cdd1dd5272 Fix the promise_intr function, it should use 'lun' not 'unit' 
to get the offset into ata_devices.
 1999-04-22 08:07:44 +00:00
luoqi
6ea0e19959 Work around an egcs optimizer bug (i386). This should fix the active ftp 
hang problem. A bug report has been sent to cygnus.
 1999-04-21 21:28:01 +00:00
dfr
d46d361744 Update VERSREQ. 1999-04-21 19:50:27 +00:00
peter
2e6c779420 Fix my breakage of BRIDGE compiling option without IPFIREWALL.. 
(Note that if you have bridge compiled in and then kldload ipfw, bridge
 won't automatically use it - knowledge of ipfw/dummynet is compiled in)
 1999-04-21 18:23:00 +00:00
sada
b77bb336f9 The function msgrcv() could copy larger data than it should do 
under some circumstances.
PR:		kern/10765
Submitted by:	Yasuhito FUTATSUKI <futatuki@fureai.or.jp>
 1999-04-21 13:30:01 +00:00